How are the words assigned?

Each time we make what3words in a new language, we use an algorithm to place simpler and more commonly used words in the areas where the language is spoken.

We’ve designed our map to try to avoid placing addresses made of similar word combinations near each other (as much as possible). This means that the overwhelming proportion of similar 3 word combinations are so far apart that an error is easy to spot.

Of course, with over 57 trillion squares on our map, there are a handful of instances where potentially similar what3words addresses may be close to each other. However, the chance of these combinations appearing ambiguously in nearby areas in the UK is estimated at about 1 in 2.5 million.

Learn more about how we deal with similar-sounding words: